Rational self-interest

Hanlon's razor* says you should never attribute to malice what can be explained by stupidity**. But there's a related saying in economics: never attribute to stupidity what can be explained by rational self-interest. If people appear to be behaving stupidly, look for the underlying self-interest that you might have missed. Of course, people do behave stupidly, often en masse, so you won't always find an underlying rational self-interest, but you often will.

*Sometimes attributed to SF author Robert A. Heinlein.
**There's also Clarke's corollary: Any sufficiently advanced stupidity is indistinguishable from malice.

It is becoming a trope in liberal and progressive economics and politics that the right-wing and the Republican party are acting stupidly, irrationally, that they have become literally deranged, even at the highest level. I believe, however, that this view is problematic.

Part of the problem is that there's a huge equivocation floating around the economics profession concerning the definition of "rational self-interest"; similarly in philosophy and ordinary conversation, there's a huge equivocation about "rationality". At minimum, rational self-interest in economics means that people always act to maximize their subjectively conceived self-interest, and while they may make errors and mistakes, on the whole they act effectively to maximize their subjective self-interest.

This definition seems metaphysical and unfalsifiable: however people are actually behaving, we can hypothesize some subjective self-interest that the behavior maximizes, and we conclude they are acting in their rational self-interest. We can, however, talk about subjective self-interest separately from specific behaviors: we can actually test hypotheses about rational self-interest.

For example, if a lot of people are buying Acme razors, which cost 50% more than the objectively identical Potzrebie razor, we could make any number of competing hypotheses about subjective self-interest that's causing the behavior. Most obviously, we might have missed some property of Acme razors that really is objectively superior to Potzrebie razors: perhaps they last longer, or are more effective at shaving a man's upper lip or a woman's ankle bone. It might be that using Acme razors confers social status precisely due to its higher price. Perhaps people simply don't take the name "Potzrebie" seriously, and don't have confidence in its quality. It might be simply that people are persuaded by the advertising so they simply feel better using an Acme razor.

The last case highlights the specific, technical nature of the economic definition of rational self-interest. It's not relevant in this context that Acme razors don't objectively make a person feel better than Potztrebie razors: they don't produce a closer shave, they don't reduce the frequency of nicks and cuts, they don't last any longer, they're not any more comfortable to use. What's relevant is that a choice of behavior produces a difference in subjective feeling; people are acting "rationally" by choosing the behavior that really does produce the better subjective experience.

The first equivocation, then, in the use of rational self-interest is equivocating between the objective properties of the choice and the subjective consequences. But objective properties are never anything more than a means to an end — indeed one means among many — to produce subjective experience. It's often the case that people do make choices that correlate to objective properties: fresh food is objectively superior to spoiled food, and people almost universally prefer unspoiled food. But there are cases where the subjective experience, and thus the choice, is not correlated directly to the objective properties of the choices.

Similarly in philosophy and ordinary conversation, there's a sense of "rationality" that means concerned exclusively with objective truth, ignoring or excluding subjective factors such as emotion. Again, there's no problem with this definition per se, but it's clearly inappropriate to apply this definition to motivation, which is fundamentally subjective.

In one sense, "rationality" just means the true correlation between behavior and and subjective outcome; in another sense, "rationality" means having objectively true beliefs about the real world. If a lot of people believe that Acme razors give a closer shave than Potzrebie razors, they are acting in their rational self-interest by spending more for Acme razors, but this rational self-interest is predicated on an irrational belief, a false belief about the world. There's no contradiction: we're just using the word "rationality" to mean different things in different contexts, an ordinary and routine use of natural language. The fallacy of equivocation comes in when we pick the meaning from the "wrong" context.

The second fallacy about rationality and rational self-interest concerns the nature of the self-interest. There are subjective desires that are easy to empathize with, and desires that are more difficult to empathize with, and desires no one wants to admit they empathize with. It's easy to dismiss an odd or unusual desire as simply non-existent, or attribute some odd behavior to a false idea about reality. But some people, to use Winston Rowntree's example, just like lounging about in unconventional clothing, and so they do: there's nothing at all irrational about their behavior in either sense.